Reconstructing the molecular life history of gliomas
At the time of their clinical manifestation, the heterogeneous group of adult and pediatric gliomas carries a wide range of diverse somatic genomic alterations, ranging from somatic single-nucleotide variants to structural chromosomal rearrangements. Somatic abnormalities may have functional consequences, such as a decrease, increase or change in mRNA transcripts, and cells pay a penalty for maintaining them. These abnormalities, therefore, must provide cells with a competitive advantage to become engrained into the glioma genome. Here, we propose a model of gliomagenesis consisting of the following five consecutive phases that glioma cells have traversed prior to clinical manifestation: (I) initial growth; (II) oncogene-induced senescence; (III) stressed growth; (IV) replicative senescence/crisis; (V) immortal growth. We have integrated the findings from a large number of studies in biology and (neuro)oncology and relate somatic alterations and other results discussed in these papers to each of these five phases. Understanding the story that each glioma tells at presentation may ultimately facilitate the design of novel, more effective therapeutic approaches.
KeywordsGlioma Gliomagenesis Oncogenesis Senescence Telomerase
Gliomas encompass a very diverse group and account for the great majority of tumors originating in the parenchyma of the central nervous system (CNS) . Two larger glioma groups are recognized: so-called diffuse gliomas, characterized by extensive infiltrative growth into the surrounding CNS parenchyma, and more circumscribed (non-diffuse) gliomas such as pilocytic astrocytoma and ependymomas. Diffuse gliomas, by far the most frequent gliomas in adult patients, are traditionally classified according to their microscopic similarities with (precursors of) glial cells and then designated as diffuse astrocytomas, oligodendrogliomas or mixed gliomas/oligoastrocytomas. Additionally, a malignancy grade is assigned to these tumors based on the presence/absence of especially marked mitotic activity, florid microvascular proliferation (MVP), and necrosis [175, 233]. For over a century, such microscopic evaluation has provided the gold standard for the diagnosis of gliomas, assessment of prognosis and formed the basis for therapeutic management. However, multiple studies showed that a purely histopathologic classification suffers from considerable inter- and intraobserver variability [3, 46, 217].
As with other human cancers, the pathogenesis and molecular evolution of gliomas are often characterized by somatic chromosomal aberrations, widespread or focal copy number changes and targeted gain and loss of function events in oncogenes and tumor suppressor genes [211, 224, 225]. Various permutations of somatic alterations were shown to be associated with distinct tumor entities and differential sensitivities to treatment, such as a chromosome 1p/19q-codeletion in oligodendrogliomas conferring increased sensitivity to chemotherapy [26, 27]. In the course of the last two decades, it became increasingly clear that such molecular characteristics may provide a more robust and objective basis for subtyping of diffuse gliomas and both scientists and clinicians turned towards molecular markers to aid diagnosis [30, 34, 59, 234, 244]. Indeed, the International Society for Neuropathology—Haarlem Consensus Guidelines and the most recent edition of the WHO classification of CNS tumors (published in 2016) embraced the notion of an integrated histo-molecular classification of diffuse gliomas [130, 131].
IDH-wildtype: most of these histologically represent astrocytic tumors, a large percentage belonging to the highest malignancy grade, i.e., glioblastomas.
IDH-mutant and 1p/19q-non-codeleted: these tumors also generally have an astrocytic phenotype, but a much larger percentage is at first diagnosis histologically lower grade/WHO grade II or III.
IDH-mutant and 1p/19-codeleted: most of these are characterized by a prominent oligodendroglial phenotype of the tumor cells.
Following this strategy, the diagnosis of mixed glioma/oligoastrocytoma can be expected to largely disappear, except when additional molecular tests cannot be performed or do not provide unequivocal results; in that situation, not-otherwise-specified (NOS) should be added to the diagnosis to indicate that ideally such samples require further workup . Rare ‘dual genotype’ oligoastrocytomas have been reported that show polymorphic phenotypes with both a complete 1p/19q-codeletion component and non-codeleted component [93, 235]. Diffuse midline glioma, H3 K27M-mutant, was added in the WHO 2016 classification as a separate entity [116, 191, 242]. This type of diffuse glioma, which most often occurs in children, is by definition located in the ‘midline’ of the CNS (brainstem, thalamus, cerebellum and/or spinal cord) and considered as highly aggressive (WHO grade IV) irrespective of the malignancy grade assigned by histology . By far, the most frequent astrocytic tumors in children though are pilocytic astrocytomas that generally are more circumscribed (therefore, grouped under non-diffuse gliomas) and show an indolent, WHO grade I behavior [44, 105, 190]. Meanwhile, the transition from a purely histological to a histo-molecular classification of especially diffuse gliomas represents a paradigm shift and necessitates re-evaluation of histologic criteria used for grading and guidance of therapeutic decisions [164, 181].
A model for the temporal molecular pathogenesis of gliomas
Phase I: initial growth
The theory that cancer results from accumulation of mutations over time, in a subset of patients combined with contribution of inherited risk factors, has been around for over six decades and has been refined over the years [5, 79, 80, 111, 158, 160]. For the purpose of this review we will consider a glioma initiation event to be the first acquired (somatic) event towards developing glioma. This event should provide a competitive growth advantage, either by directly increasing proliferation or by creating the conditions in which increased proliferation may happen. Cells and their progeny characterized by such an event will be primed to outcompete neighboring cells giving rise to an initial submicroscopic tumor mass. While germline events that contribute to glioma risk precede such glioma initiation events, their incomplete penetrance suggests that they cannot be considered as causal for glioma formation and instead prime the environment for tumor formation . Furthermore, even if a glioma initiation event marks the first somatic event in the formation of a tumor it may not be responsible for initiating growth directly. Instead, this event may promote tumorigenesis indirectly via stochastic activation of oncogenes or repression of tumor suppressors. Genetic or epigenetic selection pressures will prioritize daughter cells with growth advantages over those without and daughter cells with lethal genotypes will rapidly disappear [64, 150, 245].
IDH-mutant diffuse gliomas
Mutations in IDH1/IDH2 are commonly considered to be glioma initiating. Several studies have shown that they are amongst the few alterations highly shared amongst gliomas at first presentation and their recurrences [9, 100, 204] which is explained by their presence in the cell of origin and all cells derived thereof. Comparing multiple biopsies from the same tumor, IDH mutation can be confidently detected in each tumor segment and thus fit the proposed criteria of a glioma initiation event [107, 124, 204]. In vitro experiments have demonstrated that IDH mutations alone are sufficient to reprogram the transcriptome and epigenome of normal cells to prevent these cells from entering a terminally differentiated state [134, 147, 213].
IDH dysregulation likely contributes to gliomagenesis via the accumulation of the oncogenic metabolite R(−)-2-hydroxyglutarate (2HG) [114, 176]. Wildtype IDH enzymatically converts isocitrate into α-ketoglutarate (α-KG) as part of the citric acid cycle, whereas mutant IDH metabolizes α-KG into 2HG [54, 231]. Both mutant and wildtype IDH alleles are, therefore, essential for the oncogenic function of IDH. IDH mutations in glioma result in genome-wide hypermethylation [159, 213], most likely due to effects of 2HG on the ten–eleven translocation methylcytosine dioxygenase (Tet) family of proteins [61, 134, 243]. This hypermethylation may provide a growth advantage to cancer cells due to the epigenetic activation of oncogenes via stochastic activation of alternative gene regulatory programs, some conferring added fitness . One such mechanism in glioma may be through methylation-induced disruption of a CCCTC-binding factor (CTCF) binding site, resulting in aberrant activation of platelet-derived growth factor receptor alpha (PDGRFA) .
Although most experimental models of IDH involve overexpression of mutant IDH in vitro, several transgenic mouse models have been described [28, 125]. Early transgenic models showed that conditional knock-in of mutant IDH in the murine brain led to perinatal lethality . A more recent inducible model demonstrated that mutant IDH led to increased proliferation and infiltration in the CNS parenchyma of murine neural stem cells . Though these mice eventually died due to hydrocephalus and did not develop malignant tumors, they showed symptoms of the initial phases of gliomagenesis. Thus, while experimental models of mutant IDH are generally insufficient to cause glioma, mutant IDH leads to changes that could be interpreted as early tumor development.
IDH-wildtype diffuse astrocytomas
Approximately 70% of IDH-wildtype diffuse astrocytomas are characterized at the molecular level by a single copy loss of chromosome 10 and gain of chromosome 7 (+ 7/− 10) [16, 34]. Loss or diploid loss of heterozygosity of chromosome 10 and of chromosome arm 10p in particular has been reported to occur more frequently and may precede gain of chromosome 7 . Based on evolutionary modeling using primary-recurrent tumor pairs and multisector tumor sampling, several independent groups have found that + 7/− 10 is homogeneous and longitudinally preserved and thus likely the first and glioma initiation event in a large fraction of IDH-wildtype diffuse astrocytomas/glioblastomas [69, 107, 168, 198, 229]. A recent study suggested that gains of chromosome 7 likely occur early in tumorigenesis, amongst the first 10% of driver events . In the past several years, there has been a lot of interest in the role of TERT promoter mutations in oncogenesis and an increasing body of evidence suggests that these mutations precede + 7/− 10 . Nevertheless, a potential role of TERT promoter mutations to promote proliferation in the initial growth phase and as a glioma initiation event is speculative and will be discussed later in this review.
Chromosome 7 is home to several oncogenes that have been implicated in gliomagenesis such as cyclin-dependent kinase 6 (CDK6), MET proto-oncogene (MET) and epidermal growth factor receptor (EGFR), while chromosome 10 hosts several tumor suppressor genes, including Tet family member Tet methylcytosine dioxygenase 1 (TET1) and phosphatase and tensin homolog (PTEN). Though these genes comprise the prime suspects, it is unlikely that they alone are responsible for initiating glioma development . Studies across different diseases and in various model organisms have shown that large chromosomal copy number changes led to gross gene dosage fluctuations impacting various specific and general cellular functions . Such changes may, therefore, act in concert to promote tumor development.
Some IDH-wildtype diffuse gliomas show cytogenetically intact chromosomes 7 and 10, implying that other initiation events give rise to these tumors as well. Such events may include activating or inactivating alterations in the phosphoinositide 3-kinase (PI3K), receptor tyrosine kinase (RTK) and mitogen-activated protein kinase (MAPK) pathways . PI3K pathway alterations include mutations in PI3-kinase subunit alpha (PIK3CA), PI3-kinase subunit P85-alpha (PIK3R1), or inactivation the aforementioned tumor suppressor PTEN [127, 173, 199]. Glioma initiation events may also include point mutations in RTK pathway genes such as EGFR and PDGFRA or in MAPK pathway genes such as neurofibromin 1 (NF1) . Much is already known about the effect of these mutations on cancer growth but additional research is needed to secure their potential role as glioma initiation events.
A particular subgroup of diffuse IDH-wildtype gliomas is characterized by mutations in H3 histone family members and these gliomas occur most often in children [191, 202, 241]. The diffuse midline glioma, H3 K27M-mutant, shows a lysine to methionine substitution at position 27 of the H3 histone family member 3A (H3F3A) or histone cluster 1 H3 family member 3B (HIST1H3B) gene and is included in the WHO 2016 classification as a separate entity. Other H3-mutant diffuse gliomas in children and adolescents occurring predominantly in the cerebral hemispheres often show H3 G34R/V mutation (implying a glycine 34 to arginine or valine substitution) [115, 201, 202, 241]. In contrast to hypermethylated IDH-mutant gliomas, these H3-mutant gliomas have a general DNA hypomethylation phenotype . A recent study showed that expression of mutant H3 K27M in neonatal mice brains led to ectopic proliferation, indicating a possible pre-cancerous change . Although additional support is needed, combined with their apparent mutual exclusivity with mutations in IDH and changes characteristic for IDH-wildtype tumors, these findings indicate that H3 K27M and H3 G34R/V mutations may be glioma initiation events .
Recent studies have shown that pilocytic astrocytomas near universally harbor abnormalities in the MAPK pathway, and most commonly a tandem duplication targeting chromosome 7q, which gives rise to a KIAA1549–BRAF fusion gene consisting of the N terminus of KIAA1549 and the kinase domain of v-RAF murine sarcoma viral oncogene homolog B1 (BRAF) . Alternative alterations include the oncogenic V600E missense mutation also targeting BRAF . The BRAF V600E mutation results in an activating change due to a substitution of valine with glutamic acid at codon 600. In a non-cancer setting, BRAF activates kinases MEK and ERK, which in turn activate transcriptional machinery to promote differentiation, proliferation, growth and apoptosis . Both BRAF V600E mutation and BRAF fusion genes contribute to tumorigenesis by constitutively activating the kinase domain of BRAF, resulting in overactive signaling activity and a selective growth advantage for affected cells [55, 67, 102]. In most pilocytic astrocytomas (even after thorough analysis), an activating change in BRAF or other MAPK pathway members is the only genomic change that can be confidently detected, implying that it is the glioma initiation event in this disease .
Phase II: oncogene-induced senescence
In some cancer types, a senescent precursor stadium can be identified, such as intestinal polyps in colon cancer and dysplastic naevi in melanoma . In contrast to the catastrophic karyotypes demonstrated by their later stage derivatives, these growth-arrested senescent precursor lesions entered a senescent state via the activation of single oncogene such as BRAF and are otherwise genetically unremarkable [72, 117, 151, 153]. This begs the question if such premalignant precursor changes exist for diffuse gliomas as well. If so, such precursor lesions may, however, well be (sub)microscopic in size and go unnoticed in imaging or autopsy studies.
The tumor suppressor proteins p16INK4a, p14ARF, RB and p53 can be considered as gatekeepers of senescence. The INK4a/ARF locus on chromosome 9 contains both cyclin dependent kinase inhibitor 2A (CDKN2A) and 2B (CDKN2B), combined encoding for both p16INK4a and p14ARF via alternative splicing. Tumor suppressors RB and p53 on the other hand are encoded for by the genes retinoblastoma 1 (RB1) and tumor protein 53 (TP53) on chromosomes 13 and 17, respectively. Inactivation of one or multiple of these genes via genomic deletion and/or inactivating mutations has been linked to repression of senescence signaling and is a common event in all cancers including gliomas [39, 83, 195, 240]. For the purpose of this review, we define the term ‘senescence bypass event’ as any molecular alteration that suppresses the onset of oncogene-induced senescence.
IDH-mutant diffuse astrocytomas
IDH-mutant astrocytomas are often characterized by loss of one allele of TP53, combined with a loss-of-function mutation in the remaining allele. The frequency of TP53 mutations in IDH-mutant (‘secondary’) glioblastomas is comparable to that in lower grade IDH-mutant astrocytomas from which these glioblastomas are derived via malignant progression, suggesting that TP53 aberrations are early lesions in these tumors . Furthermore, and in contrast to IDH-wildtype gliomas, TP53 mutations were shared between all TP53-mutant cases of primary and recurrent tumors in a recent study . Analysis of multiple biopsies from the same IDH-mutant tumors indicated that samples mutant for TP53 were always IDH mutant, while some IDH-mutant samples lacked TP53 mutations, suggesting that IDH mutations precede TP53 inactivation . IDH mutation and TP53 inactivation, therefore, both comprise early events in gliomagenesis, with TP53 inactivation generally following mutation in IDH.
The p53 tumor suppressor protein is involved in many different functions, and especially its role in cell cycle arrest and senescence is very well understood . While enzymatically active wildtype p53 triggers senescence in response to oncogenic stress, mutant p53 inadequately blocks proliferation thereby bypassing the onset of oncogene-induced senescence. In addition, p53 takes a prominent role in the senescence pathway (Fig. 2) and loss of its enzymatic activity impacts replicative senescence, triggering crisis . This is discussed further in phase IV (“Phase IV: Replicative senescence/crisis” section).
IDH-mutant oligodendrogliomas, 1p/19q-codeleted
The majority of IDH-mutant tumors wildtype for TP53 demonstrate a combined single copy loss of the complete chromosome arms 1p and 19q (complete 1p/19q-codeletion) [20, 27]. These tumors are the canonical oligodendrogliomas according to the revised WHO criteria . Though it has been suggested that 1p/19q-codeletions are the result of an unbalanced translocation, much about the contribution of this event to oncogenesis remains to be resolved . 1p/19q-codeletion was found to be stable across longitudinal samples and multiple biopsies, suggesting that they are early events [1, 97, 218]. The finding that codeleted tumors are almost exclusively IDH-mutant, while the reverse is not true, suggests that mutations in IDH precede codeletion. This suggestion may implicate a common cell of origin for both astrocytoma (IDH-mutant, non-codeleted) and oligodendroglioma (IDH-mutant and 1p/19q-codeleted). Indeed, recent evidence was provided that substantiates this hypothesis, demonstrating that differences in inferred cell identity between oligodendrogliomas and astrocytomas may be entirely explained by different microenvironment makeup and the impact of the 1p/19q-codeletion on the expression of genes on these chromosome arms . The authors found that although these histological tumor types differ in morphology, these histological subtypes share a comparable developmental hierarchy and glial lineage. Transcription factors involved in oligodendrocyte differentiation are also expressed in both histologies . Several case reports have highlighted ‘dual genotype’ oligoastrocytomas, demonstrating molecular features of both bona fide astrocytic and bona fide oligodendroglial tumor cells [93, 235]. Such a dual genotype may be explained by assuming that very early in gliomagenesis a subset of IDH-mutant cells experiences a complete 1p/19q-codeletion and a TERT promoter mutation, while another IDH-mutant subset acquires a mutation in TP53 and/or ATRX. Detailed and sufficiently powered longitudinal studies of primary and recurrent diffuse gliomas may help to elucidate the actual frequency of such ‘dual genotypes’, how they are related to mixed histological appearances, and how they evolve over time .
Additional loss-of-function mutations in far-upstream element binding protein (FUBP1) on 1p31.1 and capicua transcriptional repressor (CIC) on 19q13.2 are observed in over 60% of 1p/19q-codeleted gliomas . A paired analysis of primary and recurrent 1p/19q-codeleted oligodendrogliomas described distinct alterations in CIC and FUBP1 in the primary and the recurrent tumor . Another report described that these events were frequently private to either primary or recurrence, but not both . Both findings corroborate observations of CIC and FUBP1 mutation heterogeneity across nine distinct samples from the same oligodendroglioma, including finding five distinct CIC mutants across nine tumor samples . Loss-of-function mutations in these genes led to a loss of protein expression and the FUBP1 mutation was associated with adverse survival compared to wildtype tumors [35, 90]. These findings suggest an important role for these genes and indicate that in cases in which these events were not found in one sample of the tumor, they still might be present elsewhere. Despite their apparent importance, the role of these events in gliomagenesis remains to be understood.
It is not clear whether 1p/19q-codeleted oligodendrogliomas undergo senescence or acquire mechanisms to bypass senescence. 1p/19q-codeleted gliomas generally lack alterations in genes associated with oncogene-induced senescence such as BRAF or senescence bypass such as TP53, RB1 or CDKN2A. Nevertheless, given their continued clinical growth it appears that these tumors have somehow evaded growth arrest and senescence barriers. Perhaps the 1p/19q-codeletion allows pre-cancerous cells to avoid senescence through the mono-allelic inactivation of tumor suppressor genes on these chromosome arms . One mechanism that was previously proposed may be via a dosage-dependent repression of chromodomain helicase DNA-binding domain 5 (CHD5) on 1p36 [7, 8, 222]. A study that used genetic engineering to create mouse models with gains and losses of a region corresponding to human 1p36 found that duplication of this region led to decreased proliferation and senescence whereas a single-copy deletion led to immortalization .
IDH-wildtype diffuse astrocytomas
Amongst IDH-wildtype astrocytomas/glioblastomas, one of the most frequent alterations is a homozygous loss of CDKN2A and CDKN2B . Mathematical modeling has suggested that homozygous CDKN2A/B loss occurs after +7/−10 but before other molecular events . Homozygous CDKN2A/B loss alone is insufficient for tumor formation in mice, requiring the activation of an oncogene to generate tumors in vivo . As such, homozygous CDKN2A/B loss is likely a second event in the tumorigenesis of IDH-wildtype astrocytoma or glioblastoma. The role of protein products p16INK4a and p14ARF in senescence is very well understood. Indeed, astrocytes with a homozygous deletion of CDKN2A/B can grow indefinitely in culture, and introduction of p16INK4a in immortal human glioma cell lines with this deletion leads to cell cycle arrest and senescence [87, 216]. Taken together, these results indicate that loss of CDKN2A/B may provide adult IDH-wildtype astrocytomas with a reliable means for senescence bypass.
Mutations in TP53 sometimes co-occur with homozygous CDKN2A/B loss in IDH-wildtype glioma . While TP53 mutations are often shared across all tumor cells in IDH-mutant astrocytoma, TP53 mutations in IDH-wildtype astrocytomas are frequently unique to one or a few tumor subclones . In this same study, it was found that amongst IDH-wildtype astrocytomas, whereas CDKN2A/B is consistently deleted, TP53 mutations are frequently lost or gained at tumor recurrence. These observations suggest that, in IDH-wildtype astrocytoma, CDKN2A/B may be primarily important for senescence regulation
Pediatric H3-mutant/IDH-wildtype diffuse gliomas are TP53 mutant in about 50% of cases, which may act as a senescence bypass event in these tumors . Amongst remaining H3-mutant tumors, approximately 20–30% of H3 K27 mutant diffuse gliomas show mutations in Activin A Receptor Type 1 (ACVR1) [23, 65, 208, 242]. There is no evidence that ACVR1 has any role in sentence regulation; however, ACVR1 alterations were found to be mostly mutually exclusive with alterations in TP53 and PPM1D. PPM1D is a protein phosphatase downstream of p53 involved in apoptosis regulation following DNA damage and thus likely involved in senescence . Expression of ACVR1 mutants in Tp53null murine astrocytes implanted in mouse brains failed to induce tumors, likely because H3 K27 mutations are required as a tumor initiation event and thus suggesting that ACVR1 does not meet the criteria for a tumor initiation event . Thus, while about half of pediatric H3-mutant gliomas demonstrate TP53 mutations that may act to bypass senescence, it remains unclear if and how TP53-wildtype H3-mutant pediatric gliomas bypass senescence.
Several lines of evidence suggest that pilocytic astrocytomas (WHO grade I, IDH-wildtype) are arrested in a senescent phase II state and do not advance to later phases. First, these tumors were found to frequently demonstrate several biomarkers of senescence at tumor detection, including widespread β-galactosidase activity and p16INK4a staining [88, 112, 177]. Second, these tumors are very quiescent genetically, often demonstrating but a single activated oncogene, such as a BRAF fusion, BRAF V600E mutation or rarely an activating mutation in FGFR1 or PTPN11 [101, 212, 246]. Third, these tumors generally grow slowly, have excellent outcomes and sometimes regress, perhaps because the tumor cells do not immortalize [24, 25, 75, 185]. Fourth, expression of activated BRAF V600E alone does not lead to tumor development in in vitro and in in vivo mouse models, while the combined activation of BRAF and loss of CDKN2A/B is transforming, suggesting that additional mutations to bypass senescence are required to advance to phase III [92, 182, 189]. The clinical presentation of pilocytic astrocytoma may be the result of senescence-mitigating circumstances, such as a cell of origin with proliferative potential in the absence of senescence-bypass.
Phase III: stressed growth
Cells presenting with continued proliferative signaling beyond the oncogene-induced senescence barrier are generally characterized by defective DNA damage response signaling and continued growth in a stressed environment. During this phase the repetitive DNA at the telomeric terminal ends of chromosomes become increasingly important . Telomeres progressively shorten as cells divide due to the linear conformation of chromosomes and directional replication machinery, a phenomenon that is critically important for diseases like cancer which are characterized by often rampant proliferation . Telomeric DNA takes on a lasso conformation called the t-loop, and these loops are bound by the shelterin DNA-binding protein complex. Together, these characteristics protect chromosome ends from being recognized as DNA double-strand breaks and prevent inadvertent activation of DNA damage response pathways .
The intensity of genomic instability endured during the stressed growth phase may depend on the severity of senescence pathway dysregulation incurred overcoming oncogene-induced senescence (Fig. 2). In the case of H3-mutant, IDH-wildtype and IDH-mutant astrocytoma/glioblastoma this pathway is perturbed close to the source via the direct loss of RB, p53 or p16INK4a protein function. In IDH-mutant and 1p/19q-codeleted oligodendroglioma this pathway may be repressed indirectly, for example, via the modulation of p14ARF activity through a partial deletion of CHD5. This may explain why the latter group of tumors shows significantly less genomic instability compared to the former tumor types. Moreover, in pilocytic astrocytoma this pathway may not be affected at all and these tumors may not advance beyond oncogene-induced senescence. While genomic instability incurred during phase III demonstrates some patterns that are unique to a certain glioma subtype, these features are generally shared across all gliomas regardless of subtype. We, therefore, did not separate this section according to tumor type as we did for the other phases. More research is needed to carefully delineate the selective pressures at play to better understand differences and similarities between various glioma entities in this respect.
The advent of high-throughput sequencing has led to remarkable progress in understanding the complexity of genomic instability in cancer, including complex deletions, amplifications and translocations . It is important to note that the genomic organization that can be reconstructed using sequencing at the time of analysis are those changes that resulted in viable cells and were selected for. Recent work has shown that telomere dysfunction directly leads to catastrophic genomic events, including genome shattering (chromothripsis), clustered regions of focal hypermutation (kataegis) and whole genome doubling (tetraploidization) [60, 137, 138]. Although the incidence of chromothripsis across the spectrum of gliomas is not known, a recent report suggests that chromothripsis is very common in glioblastoma . Comparison of primary and recurrent tumors across various tumor types including gliomas demonstrated that recurrent tumors lack additional genomic instability, suggesting that these events occurred during a stressful period, the initial development of the tumor that was later stabilized .
Recently, there has been a renewed interest in circular extrachromosomal DNA elements called double minute (DM) chromosomes in cancer, and it was shown that such DMs are common in gliomas [57, 214]. Although DM chromosomes have been long recognized as a cytogenetic feature of cancer, relatively little was known about its biological relevance. DM chromosomes have a predisposition to involve cancer oncogenes such as MYC proto-oncogene protein (MYC), MDM2 proto-oncogene (MDM2) or cyclin-dependent kinase 4 (CDK4) [53, 113, 156, 186, 247]. A unique feature of DM is that they lack centromeres and telomeres to dictate the organization of the mitotic spindle during mitosis and are, therefore, randomly distributed across daughter cells . Interestingly, this feature hypothetically provides DM chromosomes with an impressive fitness advantage over linear chromosomes as they do not need telomeres to protect them from inadvertent DNA damage response pathways and are not subjected to detrimental BFB cycles. It has been proposed that DM chromosomes are the result of the fusion and circular assembly of stretches of linear DNA consisting of highly homologous sequences of inverted duplications in tandem following sequential BFB cycles (Fig. 3b) . Others have proposed that ineffective DNA repair following chromothripsis can lead to linear DNA fragments getting pieced together in a circular fashion (Fig. 3c) . A study by our group found evidence of chromothripsis in several glioblastoma samples localized to chromosome 12 involving MDM2 and CDK4 and suggested that these segments may be arranged in extrachromosomal DMs . Because these are inherently random processes, it is possible that DM chromosomes that promote survival are positively selected for during telomere dysfunction. Compared to IDH-mutant gliomas, DMs in IDH-wildtype tumors more often involve established glioma oncogenes, despite what appears to be a comparable frequency of DMs in both glioma categories . More research is needed to precisely determine the frequency of DMs in glioma subtypes and to pinpoint the genetic origin of these structures.
A recent study of paired primary and recurrent IDH-mutant gliomas reported that in some cases allelic imbalances of the IDH-mutant allele occurred upon recurrence, which led to a change in mutant protein expression and consequently decreased 2HG production . Furthermore, IDH-mutant tumors are very hard to culture, and when it succeeds, IDH mutations that were present initially have been reported missing, raising the possibility that losing an IDH mutation is advantageous for survival in culture . Introduction of mutant IDH in cell cycle checkpoint-deficient cells rapidly transforms these cells into competent tumor cells . However, IDH inhibition in these cells after as little as 4 days after its first introduction did little to slow tumor growth. These findings suggest that some IDH-mutant gliomas may rapidly evolve and acquire additional driver events to uphold the tumor cell population.
To summarize, telomere dysfunction and stressed growth may promote the context-dependent evolution of glioma cells, sometimes rendering glioma initiation events redundant and providing gliomas with new fuel that rapidly increase intratumoral heterogeneity and can deal with various toxic stresses and bottlenecks. While stochastic mutational mechanisms in the stressed growth phase provide ample selection pressure to acquire beneficial changes, the detrimental genomic instability under which cells must operate acts as a powerful tumor suppressive barrier. Unchecked growth will rapidly lead to another round of DNA damage-induced replicative senescence, or when those checkpoints fail completely, cell crisis.
Phase IV: replicative senescence/crisis
Sustained stressed growth is not durable and will eventually lead tumor cells down to one of two possible roads. Tumor cells with a partially intact senescence response (i.e., functional p53 and RB) may undergo a second round of senescence called replicative senescence in response to dysfunctional telomeres. Tumor cells with a completely dysfunctional senescence response (i.e., loss-of-function mutation in TP53 or RB1) instead continue proliferating in a state of cellular crisis leading to cell death in a vast majority of cells . Replicative senescence and crisis both pose a second population bottleneck to further tumor formation. It is essential that tumor cells transition to a less stressful environment with proper telomere maintenance to prevent further BFB cycles and other catastrophic events. Cell culture experiments have demonstrated that direct immortalization of cells prior to a stressed growth phase enables them to bypass genomic instability and immortalize lacking the wild karyotypes typically associated with malignant transformation [68, 154]. These observations suggest that genomic instability in cancer development generally precedes immortal growth and is required to generate errors enabling telomere maintenance .
Acquisition of a telomere maintenance mechanism endows cancer cells with immortal growth, meaning that they are bestowed a limitless replicative potential . Telomere maintenance is established once a tumor cell has reactivated telomerase or activated alternative lengthening of telomeres. Moreover, restoration of telomere function may prevent further BFB cycles and restore genome stability. The canonical pathway involves the reactivation of the ribonucleoprotein telomerase which is transcriptionally silent in differentiated adult cells . The telomerase catalytic component telomerase reverse transcriptase (TERT) is expressed in over 80% of human cancers and is thought to be rate limiting for telomerase activity . In the alternative pathway tumors become immortalized via a recombination-driven mechanism called alternative lengthening of telomeres (ALT) .
IDH-mutant diffuse astrocytomas
IDH-mutant diffuse astrocytomas almost universally demonstrate ALT . ALT cells present with several defining characteristics, including a heterogeneous distribution of telomere length across chromosomes, extrachromosomal telomeric DNA fragments in a circular configuration (c-circles), increased expression of telomeric repeat-containing RNA (TERRA) from telomeres, the formation of ALT-associated promyelocytic leukemia bodies (APBs), frequent telomere sister chromatid exchanges (T-SCEs) and recombination between telomeres from different chromosomes . Telomeres in ALT cells are heterogeneous in length and relatively long, demonstrating telomere lengths much longer than telomerase-positive cells on average . ALT provides cancer cells with stabilizing telomere maintenance in a telomerase-negative setting. Although it remains unclear how ALT becomes activated, its presence has been tightly associated with loss-of-function events targeting the α-thalassemia/mental retardation syndrome X-linked (ATRX) or death-domain-associated protein (DAXX) genes and these events are also a hallmark feature of IDH-mutant astrocytomas [84, 98, 104]. ATRX functions as an ATP-dependent helicase within the SWI/SNF family and combined these two genes form the ATRX–DAXX complex, which functions as a histone chaperone to deposit the histone variant H3.3 at telomeres . Telomeric DNA has a tendency to form secondary quadruplex structures that challenge the replication machinery and need to be resolved for proper replication . How exactly these two genes protect telomeres from recombination and ALT is still unknown. It has been suggested that the combined helicase activity of ATRX and the histone chaperone capabilities of the ATRX–DAXX complex can resolve the secondary quadruplex structure at telomeres, thereby enabling proper progression of the replication fork during S-phase and preventing the inadvertent activation of recombination (ALT) mechanisms [121, 122].
The presence or absence of inactivating ATRX and DAXX mutations present a strong correlation with ALT in many tumor types including gliomas [84, 133, 191]. Recent in vitro studies have shown that knockout of ATRX alone is insufficient to cause ALT; however, ATRX knockout combined with inactivation of p53 and RB enzymatic activity led to an increased incidence of ALT after enduring several cycles of telomere induced crisis [40, 155, 180]. Furthermore, the reintroduction of ATRX expression in ATRX mutant ALT cells led to a repression of T-SCE, APBs and c-circle formation [40, 155].
IDH-mutant oligodendrogliomas, 1p/19q-codeleted
Oligodendrogliomas, IDH-mutant and 1p/19q-codeleted almost universally use telomerase to maintain telomeres and virtually all of these tumors carry an activating TERT promoter mutation . These TERT promoter mutations are amongst the most common non-coding mutations in cancer [89, 91, 106, 223]. Recurrent hotspot point mutations substitute a cytosine at − 228 or − 250 relative to the promoter to a thymine (C228 > T or C250 > T) to create a de novo e-twenty-six (ETS) transcription factor binding site that recruits the ETS family member GA-binding protein alpha chain (GABPA) to activate transcription .
Although the timing of TERT promoter mutations is still under debate, several lines of evidence suggest that TERT promoter mutations arise early in gliomagenesis and perhaps even occur prior to the glioma initiation event. TERT promoter mutations preferentially occur in tissues with a lower rate of self-renewal and there are numerous reports on the extra-telomeric functions of TERT, including effects on the NF-κB and WNT/β-catenin pathway promoting tumor growth and invasiveness [106, 141, 169]. Combined, this raises the possibility that these mutations may contribute to tumorigenesis via other pathways than its effect on telomerase alone, providing a biological reason for these mutations to contribute to gliomagenesis early in phase I and before the onset of dysfunctional telomeres. In a glioma-specific analysis, it was found that nearly all tumors with the ‘phase I event’ + 7/− 10 or ‘phase II event’ 1p/19q-codeletion have TERT promoter mutations, whereas not all TERT promoter mutant gliomas have + 7/− 10 or 1p/19q-codeletions, which may indicate that TERT promoter mutations even precede + 7/− 10 and 1p/19q-codeletions . Another group studied the mutation fraction using multisector sequencing in 1p/19q-codeleted oligodendrogliomas and found that TERT promoter mutations indeed are early events and may occur before IDH mutations . The idea that TERT promoter mutation occurs early is further corroborated by the finding that genetically engineered TERT promoter mutations in telomerase-positive embryonic stem cells do not affect telomerase activity, while upon differentiation these engineered cells remain telomerase positive and acquire immortality . More recently, it was found that TERT promoter mutations in melanoma initially do not support telomere maintenance and telomeres shorten to critically short length despite harboring promoter mutations . The effect on telomere length was not observed until later, which the authors attributed to a two-step immortalization process. One study even reported canonical (− 228 and − 250) somatic TERT promoter mutations in the blood of multiple non-cancer patients, indicating that these events could even occur before the onset of cancer and act to prime the tumor bed . Although roles for TERT outside of telomere maintenance remain to be understood, these observations provide a sound argument that TERT promoter mutations can occur early in or even before gliomagenesis while providing a means towards immortalization at a later stage.
IDH-wildtype diffuse astrocytomas
The majority of IDH-wildtype diffuse gliomas use telomerase for telomere maintenance . Re-analysis of previously published samples reclassified according to WHO 2016 criteria demonstrated that approximately 75% of diffuse IDH-wildtype gliomas are TERT promoter mutant . Thus, TERT promoter mutations are common in both the most and the least aggressive diffuse gliomas (IDH-wildtype diffuse astrocytomas and IDH-mutant and 1p19q-codeleted oligodendrogliomas, respectively), suggesting that TERT promoter mutations are not dictating their biological behavior. It was further found that ATRX mutations occur in approximately 5% of IDH-wildtype diffuse astrocytomas . The prevalence of ALT in IDH-wildtype diffuse gliomas is higher than the frequency of ATRX mutations, suggesting that some of these tumors may use ATRX-independent mechanisms to activate ALT . In a similar fashion, the prevalence of telomerase activity is higher than the prevalence of TERT promoter mutations in this tumor type, suggesting that these tumors may use TERT promoter-independent mechanisms for the reactivation of telomerase . Several candidate mechanisms have been previously described in glioma, including TERT promoter methylation or TERT amplifications . Contrary to adult IDH-wildtype diffuse glioma, H3-mutant malignant pediatric glioma frequently demonstrates ALT , and several studies reported frequent co-occurrence of ATRX mutations in both H3 K27 and G34 mutant gliomas. However, reports of co-occurrence vary between 30 and 60% for K27 and 75–100% for G34, suggesting that there is a role for telomerase in many of these tumors as well .
Phase V: immortalization and dedifferentiation
The glioma stem cell theory states that amongst all cancerous cells in a tumor, a subset of cells act as progenitor or stem cells with reproductive capabilities and sustaining the cancer, much like normal bone marrow stem cells are responsible for replenishing the population of circulating leukocytes . It has often been contrasted to the theory of clonal evolution, which suggests that cancers evolve through an iterative process of clonal expansion from a single cell . Recent advances in single-cell sequencing and lineage tracing have unveiled multiple populations of tumor cells in bulk tumor samples, providing fuel for the cancer stem cell hypothesis [119, 171, 210, 219]. One study used single-cell RNA sequencing on IDH-mutant and 1p/19q-codeleted oligodendroglioma patient samples and uncovered distinct cell populations of undifferentiated tumor stem cells and cells that are more differentiated along various glial lineages . In a similar study of IDH-mutant astrocytoma the authors were able to detect the same cellular populations but with a higher ratio of stem-like to differentiated cells that increased with increasing WHO grades . Another study used DNA barcoding of repeatedly in vivo-transplanted glioblastoma cells to trace the lineage during their engraftment and found a population of progenitor cells that sustained the tumor and gave rise to differentiated non-proliferative cells .
Aforementioned studies supporting the cancer stem cell hypothesis suggest that multiple cellular populations exist within a tumor, including a self-renewing cancer stem cell population and a less-proliferative differentiated population. A recent paper demonstrated that cells derived from glioma stem cells may differentiate and subsequently undergo senescence . We speculate that phase V tumor cells represent cancer stem cells that may give rise to more differentiated, phase IV cells (Fig. 4a). The transition from phase IV to V and vice versa is likely volatile in nature owing to transcriptional reprogramming including the activation of the stemness factors oligodendrocyte transcription factor 2 (OLIG2), sex-determining region Y-box 2 (SOX2) and the reactivation of telomerase . On the other hand, the transition from phase I to phase IV is more rigid in nature, involving various changes on a genomic level, including somatic mutations and copy number changes as described. This important distinction in flexibility led us to believe that phase V cells may re-differentiate and assume a phase IV state.
Although they are growth arrested, senescent cells are metabolically active and release a plethora of signaling molecules to the microenvironment. The senescence-associated secretory phenotype (SASP) is a feature of senescent cells that curtails these cells with the release of proinflammatory cytokines . SASP does not depend on p16INK4a or p21 activity and senescence with intact p16INKa function actually suppresses SASP . Similarly, activated p53 signaling also suppresses SASP while TP53 loss induces SASP [48, 136, 178]. These findings suggest that the SASP response is stronger when senescence pathway genes are lost. Indeed, IDH-wildtype astrocytomas often harbor homozygous deletions in CDKN2A/B and are known to have a highly active microenvironment . Genes associated with SASP were shown to be overexpressed in higher grades of glioma and older patients, the latter group more likely to be affected by high-grade IDH-wildtype astrocytoma . Moreover, it was found that primary glioblastoma cells retain a functional senescence program despite mutations in the TERT promoter and CDKN2A/B locus . These findings imply that senescent and differentiated phase IV cells may be crucial for shaping the immune microenvironment in gliomas (Fig. 4b).
A broader perspective
While there are not many known environmental risk factors that predispose to glioma, large cohort genome-wide association studies over the past two decades have identified multiple heritable polymorphisms conferring glioma risk [96, 110, 152, 179, 196, 238]. Notably, several of these risk loci are localized to genes involved in telomere maintenance, including the telomerase reverse transcriptase TERT, telomerase RNA component TERC, and other telomere maintenance-associated genes STN1, CST complex subunit (OBFC1), protection of telomeres 1 (POT1) and regulator of telomere elongation 1 (RTEL1) . Moreover, there appears to be a significant increased glioma risk in people with increased leukocyte telomere lengths . Unsurprisingly, glioma risk alleles at aforementioned genes are also associated with increased leukocyte telomere length [41, 226, 228]. Telomeres thus play an important role in not only the development of gliomas, but also in glioma risk . In fact, the positive association between leukocyte telomere length and cancer risk is not specific to glioma and shared across many cancers. A recent Mendelian randomization study found that longer leukocyte telomere length was associated with an increased risk to cancer but a reduced risk to non-neoplastic disease such as cardiovascular disease . It has been suggested that this difference is due to individuals with longer telomeres being more likely to acquire driver mutations due to an increased proliferative potential whereas the inverse relationship observed for non-neoplastic disease may be due to the impact of telomere shortening on tissue degeneration [19, 200].
Several hereditary disorders are associated with an increased risk of glioma development, including neurofibromatosis type 1 and type 2 (NF1, NF2) and the TP53 germline mutation/Li–Fraumeni syndrome. NF1 and NF2 are autosomal dominant hereditary disorders with germline mutations in NF1 and NF2 and clinically characterized by multiple benign nerve sheath tumors (especially neurofibromas in NF1, schwannomas in NF2), but also by a markedly increased risk on particular gliomas (especially pilocytic astrocytoma in NF1 and ependymomas in NF2) [33, 76, 184]. Both genes are well-known tumor suppressor genes and key components in the MAPK pathway . It has been demonstrated that senescence commonly occurs in benign nerve sheath tumors and that prolonged NF1 disruption leads to oncogene-induced senescence in a model system, providing a rationale as to why these germline disorders present with tumors that are often relatively indolent . A germline perturbation affecting NF1 or NF2 can be considered a tumor initiation event, explaining why this germline disorder guarantees the formation of multiple benign nerve sheath tumors over one’s lifetime.
Li–Fraumeni syndrome is a rare autosomal dominant hereditary disorder that is caused by the germline perturbation of TP53 or CHK2, which regulates p53 activity [12, 18, 146]. Whereas NF1 and NF2 guarantee the formation of especially multiple benign tumors (including non-diffuse gliomas) in a lifetime, Li–Fraumeni patients pose a greater risk to developing a malignant tumor, including a diffuse glioma . This risk increases with age and is over 50% at age 30, with a lifetime cancer risk of over 70% in men and almost 100% in women . Moreover, 15 and 4% of affected individuals were found to develop a second and third cancer . Li–Fraumeni syndrome germline mutations affect phase II and prevent the onset of oncogene-induced senescence following the acquisition of a glioma initiation event, thus increasing the risk of developing cancer over a lifetime.
The germline mutations underlying NF1/NF2 and Li–Fraumeni syndrome represent pathways that both need to be disrupted for a malignant tumor to form. The fact that nearly all patients with NF1/NF2 develop one or more benign tumors can be understood by acknowledging that in these disorders a germline tumor initiation event is involved. Unless this pathway is supplemented by a senescence bypass event, these tumors do not readily proceed to malignancy. In contrast, Li–Fraumeni syndrome is characterized by an increased risk for malignant tumors in many (but not all) patients. In this syndrome, the germline senescence bypass event needs to be supplemented by a tumor initiation event for a tumor to develop, and such tumors may be more aggressive/malignant due to the defective senescence barrier, allowing the tumor (precursor) cell to instantly progress to phase III and instigate genomic instability.
Our knowledge on the molecular events driving cancer has grown exponentially over the years. This review has aimed to put this new knowledge into the perspective of the temporal molecular pathogenesis of glioma, starting from the first aberrant cell all the way to a symptom-causing glioma. To this end we have combined what is known on gene (mal)function, tumor evolution, genomic instability and telomere maintenance to develop a model of gliomagenesis. This model describes five sequential phases cancer cells undergo during their gliomagenesis. We speculate that transitions from one phase to the next can be characterized by acquisition of tumor-driving events that sequentially contribute to the hallmarks of cancer as previously proposed, including proliferation, evasion of apoptosis and limitless replicative potential [79, 80]. Our model is a simplified abstraction of what may be the truth and new insights will refine and improve our understanding. Meanwhile, we hope that our model will help foster hypotheses leading to new insights into the molecular life history of glioma that will help identify convincing therapeutic vulnerabilities.
We thank members of the Verhaak laboratory for insightful discussion and valuable input. We thank Zoë Reifsnyder of the Jackson Laboratory creative team for artwork. This work was supported by Grants from the National Institutes of Health R01 CA190121 and P30CA034196, the National Brain Tumor Society Oligo Research Fund and DefeatGBM Initiative, and Grant 11026 from the Dutch Cancer Society KWF. We have no disclosures to make.
- 1.Aihara K, Mukasa A, Nagae G, Nomura M, Yamamoto S, Ueda H, Tatsuno K, Shibahara J, Takahashi M, Momose T et al (2017) Genetic and epigenetic stability of oligodendrogliomas at recurrence. Acta Neuropathol Commun 5:18. https://doi.org/10.1186/s40478-017-0422-z PubMedPubMedCentralCrossRefGoogle Scholar
- 2.Aldape K, Amin SB, Ashley DM, Barnholtz-Sloan JS, Bates AJ, Beroukhim R, Bock C, Brat DJ, Claus EB, Costello JF et al (2018) Glioma through the looking GLASS: molecular evolution of diffuse gliomas and the glioma longitudinal analysis consortium. Neurooncology. https://doi.org/10.1093/neuonc/noy020 CrossRefGoogle Scholar
- 6.Azar S, Leventoux N, Ripoll C, Rigau V, Goze C, Lorcy F, Bauchet L, Duffau H, Guichet PO, Rothhut B et al (2018) Cellular and molecular characterization of IDH1-mutated diffuse low grade gliomas reveals tumor heterogeneity and absence of EGFR/PDGFRalpha activation. Glia 66:239–255. https://doi.org/10.1002/glia.23240 PubMedCrossRefGoogle Scholar
- 10.Bardella C, Al-Dalahmah O, Krell D, Brazauskas P, Al-Qahtani K, Tomkova M, Adam J, Serres S, Lockstone H, Freeman-Mills L et al (2016) Expression of Idh 1(R132H) in the murine subventricular zone stem cell niche recapitulates features of early gliomagenesis. Cancer Cell 30:578–594. https://doi.org/10.1016/j.ccell.2016.08.017 PubMedPubMedCentralCrossRefGoogle Scholar
- 13.Bell RJ, Rube HT, Kreig A, Mancini A, Fouse SD, Nagarajan RP, Choi S, Hong C, He D, Pekmezci M et al (2015) Cancer. The transcription factor GABP selectively binds and activates the mutant TERT promoter in cancer. Science 348:1036–1039. https://doi.org/10.1126/science.aab0015 PubMedPubMedCentralCrossRefGoogle Scholar
- 14.Bender S, Tang Y, Lindroth AM, Hovestadt V, Jones DT, Kool M, Zapatka M, Northcott PA, Sturm D, Wang W et al (2013) Reduced H3K27me3 and DNA hypomethylation are major drivers of gene expression in K27M mutant pediatric high-grade gliomas. Cancer Cell 24:660–672. https://doi.org/10.1016/j.ccr.2013.10.006 PubMedCrossRefGoogle Scholar
- 16.Bigner SH, Mark J, Burger PC, Mahaley MS Jr, Bullard DE, Muhlbaier LH, Bigner DD (1988) Specific chromosomal abnormalities in malignant human gliomas. Can Res 48:405–411Google Scholar
- 18.Birch JM, Hartley AL, Tricker KJ, Prosser J, Condie A, Kelsey AM, Harris M, Jones PH, Binchy A, Crowther D et al (1994) Prevalence and diversity of constitutional mutations in the p53 gene among 21 Li-Fraumeni families. Can Res 54:1298–1304Google Scholar
- 23.Buczkowicz P, Hoeman C, Rakopoulos P, Pajovic S, Letourneau L, Dzamba M, Morrison A, Lewis P, Bouffet E, Bartels U et al (2014) Genomic analysis of diffuse intrinsic pontine gliomas identifies three molecular subgroups and recurrent activating ACVR1 mutations. Nat Genet 46:451–456. https://doi.org/10.1038/ng.2936 PubMedPubMedCentralCrossRefGoogle Scholar
- 25.Burkhard C, Di Patre PL, Schuler D, Schuler G, Yasargil MG, Yonekawa Y, Lutolf UM, Kleihues P, Ohgaki H (2003) A population-based study of the incidence and survival rates in patients with pilocytic astrocytoma. J Neurosurg 98:1170–1174. https://doi.org/10.3171/jns.2003.98.6.1170 PubMedCrossRefGoogle Scholar
- 26.Cairncross G, Wang M, Shaw E, Jenkins R, Brachman D, Buckner J, Fink K, Souhami L, Laperriere N, Curran W et al (2013) Phase III trial of chemoradiotherapy for anaplastic oligodendroglioma: long-term results of RTOG 9402. J Clin Oncol 31:337–343. https://doi.org/10.1200/JCO.2012.43.2674 PubMedCrossRefGoogle Scholar
- 27.Cairncross JG, Ueki K, Zlatescu MC, Lisle DK, Finkelstein DM, Hammond RR, Silver JS, Stark PC, Macdonald DR, Ino Y et al (1998) Specific genetic predictors of chemotherapeutic response and survival in patients with anaplastic oligodendrogliomas. J Natl Cancer Inst 90:1473–1479PubMedCrossRefGoogle Scholar
- 30.Cancer Genome Atlas Research N, Brat DJ, Verhaak RG, Aldape KD, Yung WK, Salama SR, Cooper LA, Rheinbay E, Miller CR, Vitucci Met al (2015) Comprehensive, integrative genomic analysis of diffuse lower-grade gliomas. New Engl J Med 372:2481–2498. https://doi.org/10.1056/nejmoa1402121
- 32.Castel D, Philippe C, Calmon R, Le Dret L, Truffaux N, Boddaert N, Pages M, Taylor KR, Saulnier P, Lacroix L et al (2015) Histone H3F3A and HIST1H3B K27M mutations define two subgroups of diffuse intrinsic pontine gliomas with different prognosis and phenotypes. Acta Neuropathol 130:815–827. https://doi.org/10.1007/s00401-015-1478-0 PubMedPubMedCentralCrossRefGoogle Scholar
- 34.Ceccarelli M, Barthel FP, Malta TM, Sabedot TS, Salama SR, Murray BA, Morozova O, Newton Y, Radenbaugh A, Pagnotta SM et al (2016) Molecular profiling reveals biologically discrete subsets and pathways of progression in diffuse glioma. Cell 164:550–563. https://doi.org/10.1016/j.cell.2015.12.028 PubMedPubMedCentralCrossRefGoogle Scholar
- 37.Chiba K, Lorbeer FK, Shain AH, McSwiggen DT, Schruf E, Oh A, Ryu J, Darzacq X, Bastian BC, Hockemeyer D (2017) Mutations in the promoter of the telomerase gene TERT contribute to tumorigenesis by a two-step mechanism. Science 357:1416–1420. https://doi.org/10.1126/science.aao0535 PubMedCrossRefPubMedCentralGoogle Scholar
- 41.Codd V, Nelson CP, Albrecht E, Mangino M, Deelen J, Buxton JL, Hottenga JJ, Fischer K, Esko T, Surakka I et al (2013) Identification of seven loci affecting mean telomere length and their association with disease. Nat Genet 45:422–427. https://doi.org/10.1038/ng.2528 (e421–422)PubMedPubMedCentralCrossRefGoogle Scholar
- 47.Coppe JP, Desprez PY, Krtolica A, Campisi J (2010) The senescence-associated secretory phenotype: the dark side of tumor suppression. Annu Rev Pathol 5:99–118. https://doi.org/10.1146/annurev-pathol-121808-102144 PubMedPubMedCentralCrossRefGoogle Scholar
- 48.Coppe JP, Patil CK, Rodier F, Sun Y, Munoz DP, Goldstein J, Nelson PS, Desprez PY, Campisi J (2008) Senescence-associated secretory phenotypes reveal cell-nonautonomous functions of oncogenic RAS and the p53 tumor suppressor. PLoS Biol 6:2853–2868. https://doi.org/10.1371/journal.pbio.0060301 PubMedCrossRefGoogle Scholar
- 49.Coppe JP, Rodier F, Patil CK, Freund A, Desprez PY, Campisi J (2011) Tumor suppressor and aging biomarker p16(INK4a) induces cellular senescence without the associated inflammatory secretory phenotype. J Biol Chem 286:36396–36403. https://doi.org/10.1074/jbc.M111.257071 PubMedPubMedCentralCrossRefGoogle Scholar
- 50.Coppola D, Balducci L, Chen DT, Loboda A, Nebozhyn M, Staller A, Fulp WJ, Dalton W, Yeatman T, Brem S (2014) Senescence-associated-gene signature identifies genes linked to age, prognosis, and progression of human gliomas. J Geriatr Oncol 5:389–399. https://doi.org/10.1016/j.jgo.2014.08.003 PubMedCrossRefGoogle Scholar
- 52.Courtois-Cox S, Genther Williams SM, Reczek EE, Johnson BW, McGillicuddy LT, Johannessen CM, Hollstein PE, MacCollin M, Cichowski K (2006) A negative feedback signaling network underlies oncogene-induced senescence. Cancer Cell 10:459–472. https://doi.org/10.1016/j.ccr.2006.10.003 PubMedPubMedCentralCrossRefGoogle Scholar
- 57.de Carvalho AC, Kim H, Poisson LM, Winn ME, Mueller C, Cherba D, Koeman J, Seth S, Protopopov A, Felicella M et al (2017) Discordant inheritance of chromosomal and extrachromosomal DNA elements contributes to dynamic disease evolution in glioblastoma. bioRxiv: https://doi.org/10.1101/081158
- 59.Eckel-Passow JE, Lachance DH, Molinaro AM, Walsh KM, Decker PA, Sicotte H, Pekmezci M, Rice T, Kosel ML, Smirnov IV et al (2015) Glioma groups based on 1p/19q, IDH, and TERT promoter mutations in tumors. New Engl J Med 372:2499–2508. https://doi.org/10.1056/NEJMoa1407279 PubMedPubMedCentralCrossRefGoogle Scholar
- 61.Figueroa ME, Abdel-Wahab O, Lu C, Ward PS, Patel J, Shih A, Li Y, Bhagwat N, Vasanthakumar A, Fernandez HF et al (2010) Leukemic IDH1 and IDH2 mutations result in a hypermethylation phenotype, disrupt TET2 function, and impair hematopoietic differentiation. Cancer Cell 18:553–567. https://doi.org/10.1016/j.ccr.2010.11.015 PubMedPubMedCentralCrossRefGoogle Scholar
- 65.Fontebasso AM, Papillon-Cavanagh S, Schwartzentruber J, Nikbakht H, Gerges N, Fiset PO, Bechet D, Faury D, De Jay N, Ramkissoon LA et al (2014) Recurrent somatic mutations in ACVR1 in pediatric midline high-grade astrocytoma. Nat Genet 46:462–466. https://doi.org/10.1038/ng.2950 PubMedPubMedCentralCrossRefGoogle Scholar
- 67.Forshew T, Tatevossian RG, Lawson AR, Ma J, Neale G, Ogunkolade BW, Jones TA, Aarum J, Dalton J, Bailey S et al (2009) Activation of the ERK/MAPK pathway: a signature genetic defect in posterior fossa pilocytic astrocytomas. J Pathol 218:172–181. https://doi.org/10.1002/path.2558 PubMedCrossRefGoogle Scholar
- 68.Garbe JC, Vrba L, Sputova K, Fuchs L, Novak P, Brothman AR, Jackson M, Chin K, LaBarge MA, Watts G et al (2014) Immortalization of normal human mammary epithelial cells in two steps by direct targeting of senescence barriers does not require gross genomic alterations. Cell cycle (Georgetown, Tex) 13:3423–3435. https://doi.org/10.4161/15384101.2014.954456
- 69.Gerstung M, Jolly C, Leshchiner I, Dentro SC, Gonzalez S, Mitchell TJ, Rubanova Y, Anur P, Rosebrock D, Yu K et al (2017) The evolutionary history of 2,658 cancers. bioRxiv: https://doi.org/10.1101/161562
- 70.Gisselsson D, Pettersson L, Hoglund M, Heidenblad M, Gorunova L, Wiegant J, Mertens F, Dal Cin P, Mitelman F, Mandahl N (2000) Chromosomal breakage-fusion-bridge events cause genetic intratumor heterogeneity. Proc Natl Acad Sci USA 97:5357–5362. https://doi.org/10.1073/pnas.090013497 PubMedPubMedCentralCrossRefGoogle Scholar
- 71.Goldberg AD, Banaszynski LA, Noh KM, Lewis PW, Elsaesser SJ, Stadler S, Dewell S, Law M, Guo X, Li X et al (2010) Distinct factors control histone variant H3.3 localization at specific genomic regions. Cell 140:678–691. https://doi.org/10.1016/j.cell.2010.01.003 PubMedPubMedCentralCrossRefGoogle Scholar
- 85.Heaphy CM, Subhawong AP, Hong SM, Goggins MG, Montgomery EA, Gabrielson E, Netto GJ, Epstein JI, Lotan TL, Westra WH et al (2011) Prevalence of the alternative lengthening of telomeres telomere maintenance mechanism in human cancer subtypes. Am J Pathol 179:1608–1615. https://doi.org/10.1016/j.ajpath.2011.06.018 PubMedPubMedCentralCrossRefGoogle Scholar
- 92.Huillard E, Hashizume R, Phillips JJ, Griveau A, Ihrie RA, Aoki Y, Nicolaides T, Perry A, Waldman T, McMahon M et al (2012) Cooperative interactions of BRAFV600E kinase and CDKN2A locus deficiency in pediatric malignant astrocytoma as a basis for rational therapy. Proc Natl Acad Sci USA 109:8710–8715. https://doi.org/10.1073/pnas.1117255109 PubMedPubMedCentralCrossRefGoogle Scholar
- 95.Jenkins RB, Blair H, Ballman KV, Giannini C, Arusell RM, Law M, Flynn H, Passe S, Felten S, Brown PD et al (2006) A t(1;19)(q10;p10) mediates the combined deletions of 1p and 19q and predicts a better prognosis of patients with oligodendroglioma. Can Res 66:9852–9861. https://doi.org/10.1158/0008-5472.can-06-1796 CrossRefGoogle Scholar
- 96.Jenkins RB, Xiao Y, Sicotte H, Decker PA, Kollmeyer TM, Hansen HM, Kosel ML, Zheng S, Walsh KM, Rice T et al (2012) A low-frequency variant at 8q24.21 is strongly associated with risk of oligodendroglial tumors and astrocytomas with IDH1 or IDH2 mutation. Nat Genet 44:1122–1125. https://doi.org/10.1038/ng.2388 PubMedPubMedCentralCrossRefGoogle Scholar
- 97.Jeuken JW, Sijben A, Bleeker FE, Boots-Sprenger SH, Rijntjes J, Gijtenbeek JM, Mueller W, Wesseling P (2011) The nature and timing of specific copy number changes in the course of molecular progression in diffuse gliomas: further elucidation of their genetic “life story”. Brain Pathol 21:308–320. https://doi.org/10.1111/j.1750-3639.2010.00447.x PubMedCrossRefGoogle Scholar
- 98.Jiao Y, Killela PJ, Reitman ZJ, Rasheed AB, Heaphy CM, de Wilde RF, Rodriguez FJ, Rosemberg S, Oba-Shinjo SM, Nagahashi Marie SK et al (2012) Frequent ATRX, CIC, FUBP1 and IDH1 mutations refine the classification of malignant gliomas. Oncotarget 3:709–722. https://doi.org/10.18632/oncotarget.588 PubMedPubMedCentralCrossRefGoogle Scholar
- 104.Kannan K, Inagaki A, Silber J, Gorovets D, Zhang J, Kastenhuber ER, Heguy A, Petrini JH, Chan TA, Huse JT (2012) Whole-exome sequencing identifies ATRX mutation as a key molecular determinant in lower-grade glioma. Oncotarget 3:1194–1203. https://doi.org/10.18632/oncotarget.689 PubMedPubMedCentralCrossRefGoogle Scholar
- 106.Killela PJ, Reitman ZJ, Jiao Y, Bettegowda C, Agrawal N, Diaz LA Jr, Friedman AH, Friedman H, Gallia GL, Giovanella BC et al (2013) TERT promoter mutations occur frequently in gliomas and a subset of tumors derived from cells with low rates of self-renewal. Proc Natl Acad Sci USA 110:6021–6026. https://doi.org/10.1073/pnas.1303607110 PubMedPubMedCentralCrossRefGoogle Scholar
- 107.Kim H, Zheng S, Amini SS, Virk SM, Mikkelsen T, Brat DJ, Grimsby J, Sougnez C, Muller F, Hu J et al (2015) Whole-genome and multisector exome sequencing of primary and post-treatment glioblastoma reveals patterns of tumor evolution. Genome Res 25:316–327. https://doi.org/10.1101/gr.180612.114 PubMedPubMedCentralCrossRefGoogle Scholar
- 110.Kinnersley B, Labussiere M, Holroyd A, Di Stefano AL, Broderick P, Vijayakrishnan J, Mokhtari K, Delattre JY, Gousias K, Schramm J et al (2015) Genome-wide association study identifies multiple susceptibility loci for glioma. Nat Commun 6:8559. https://doi.org/10.1038/ncomms9559 PubMedPubMedCentralCrossRefGoogle Scholar
- 112.Koelsche C, Wohrer A, Jeibmann A, Schittenhelm J, Schindler G, Preusser M, Lasitschka F, von Deimling A, Capper D (2013) Mutant BRAF V600E protein in ganglioglioma is predominantly expressed by neuronal tumor cells. Acta Neuropathol 125:891–900. https://doi.org/10.1007/s00401-013-1100-2 PubMedCrossRefGoogle Scholar
- 115.Korshunov A, Capper D, Reuss D, Schrimpf D, Ryzhova M, Hovestadt V, Sturm D, Meyer J, Jones C, Zheludkova O et al (2016) Histologically distinct neuroepithelial tumors with histone 3 G34 mutation are molecularly similar and comprise a single nosologic entity. Acta Neuropathol 131:137–146. https://doi.org/10.1007/s00401-015-1493-1 PubMedCrossRefGoogle Scholar
- 116.Korshunov A, Ryzhova M, Hovestadt V, Bender S, Sturm D, Capper D, Meyer J, Schrimpf D, Kool M, Northcott PA et al (2015) Integrated analysis of pediatric glioblastoma reveals a subset of biologically favorable tumors with associated molecular prognostic markers. Acta Neuropathol 129:669–678. https://doi.org/10.1007/s00401-015-1405-4 PubMedCrossRefGoogle Scholar
- 121.Law MJ, Lower KM, Voon HP, Hughes JR, Garrick D, Viprakasit V, Mitson M, De Gobbi M, Marra M, Morris A et al (2010) ATR-X syndrome protein targets tandem repeats and influences allele-specific expression in a size-dependent manner. Cell 143:367–378. https://doi.org/10.1016/j.cell.2010.09.023 PubMedCrossRefGoogle Scholar
- 126.Lewis PW, Muller MM, Koletsky MS, Cordero F, Lin S, Banaszynski LA, Garcia BA, Muir TW, Becher OJ, Allis CD (2013) Inhibition of PRC2 activity by a gain-of-function H3 mutation found in pediatric glioblastoma. Science 340:857–861. https://doi.org/10.1126/science.1232245 PubMedPubMedCentralCrossRefGoogle Scholar
- 129.Louis DN, Giannini C, Capper D, Paulus W, Figarella-Branger D, Lopes MB, Batchelor TT, Cairncross JG, van den Bent M, Wick W et al (2018) cIMPACT-NOW update 2: diagnostic clarifications for diffuse midline glioma, H3 K27M-mutant and diffuse astrocytoma/anaplastic astrocytoma, IDH-mutant. Acta Neuropathol. https://doi.org/10.1007/s00401-018-1826-y
- 130.Louis DN, Ohgaki H, Wiestler OD, Cavenee WK, Ellison DW, Figarella-Branger D, Perry A, Reifenberger G, Deimling AV (2016) WHO classification of tumours of the central nervous system. IARC Press, LyonGoogle Scholar
- 131.Louis DN, Perry A, Burger P, Ellison DW, Reifenberger G, von Deimling A, Aldape K, Brat D, Collins VP, Eberhart C et al (2014) International Society Of Neuropathology-Haarlem consensus guidelines for nervous system tumor classification and grading. Brain Pathol 24:429–435. https://doi.org/10.1111/bpa.12171 PubMedCrossRefGoogle Scholar
- 132.Louis DN, Wesseling P, Paulus W, Giannini C, Batchelor TT, Cairncross JG, Capper D, Figarella-Branger D, Lopes MB, Wick W et al (2018) cIMPACT-NOW update 1: not otherwise specified (NOS) and not elsewhere classified (NEC). Acta Neuropathol 135:481–484. https://doi.org/10.1007/s00401-018-1808-0 PubMedCrossRefGoogle Scholar
- 133.Lovejoy CA, Li W, Reisenweber S, Thongthip S, Bruno J, de Lange T, De S, Petrini JH, Sung PA, Jasin M et al (2012) Loss of ATRX, genome instability, and an altered DNA damage response are hallmarks of the alternative lengthening of telomeres pathway. PLoS Genet 8:e1002772. https://doi.org/10.1371/journal.pgen.1002772 PubMedPubMedCentralCrossRefGoogle Scholar
- 134.Lu C, Ward PS, Kapoor GS, Rohle D, Turcan S, Abdel-Wahab O, Edwards CR, Khanin R, Figueroa ME, Melnick A et al (2012) IDH mutation impairs histone demethylation and results in a block to cell differentiation. Nature 483:474–478. https://doi.org/10.1038/nature10860 PubMedPubMedCentralCrossRefGoogle Scholar
- 139.Malhotra A, Lindberg M, Faust GG, Leibowitz ML, Clark RA, Layer RM, Quinlan AR, Hall IM (2013) Breakpoint profiling of 64 cancer genomes reveals numerous complex rearrangements spawned by homology-independent mechanisms. Genome Res 23:762–776. https://doi.org/10.1101/gr.143677.112 PubMedPubMedCentralCrossRefGoogle Scholar
- 140.Mangerel J, Price A, Castelo-Branco P, Brzezinski J, Buczkowicz P, Rakopoulos P, Merino D, Baskin B, Wasserman J, Mistry M et al (2014) Alternative lengthening of telomeres is enriched in, and impacts survival of TP53 mutant pediatric malignant brain tumors. Acta Neuropathol 128:853–862. https://doi.org/10.1007/s00401-014-1348-1 PubMedCrossRefGoogle Scholar
- 145.Mazor T, Chesnelong C, Pankov A, Jalbert LE, Hong C, Hayes J, Smirnov IV, Marshall R, Souza CF, Shen Y et al (2017) Clonal expansion and epigenetic reprogramming following deletion or amplification of mutant IDH1. Proc Natl Acad Sci USA 114:10743–10748. https://doi.org/10.1073/pnas.1708914114 PubMedPubMedCentralCrossRefGoogle Scholar
- 151.Meeker AK, Hicks JL, Iacobuzio-Donahue CA, Montgomery EA, Westra WH, Chan TY, Ronnett BM, De Marzo AM (2004) Telomere length abnormalities occur early in the initiation of epithelial carcinogenesis. Clin Cancer Res 10:3317–3326. https://doi.org/10.1158/1078-0432.CCR-0984-03 PubMedCrossRefGoogle Scholar
- 152.Melin BS, Barnholtz-Sloan JS, Wrensch MR, Johansen C, Il’yasova D, Kinnersley B, Ostrom QT, Labreche K, Chen Y, Armstrong G et al (2017) Genome-wide association study of glioma subtypes identifies specific differences in genetic susceptibility to glioblastoma and non-glioblastoma tumors. Nat Genet 49:789–794. https://doi.org/10.1038/ng.3823 PubMedPubMedCentralCrossRefGoogle Scholar
- 156.Nikolaev S, Santoni F, Garieri M, Makrythanasis P, Falconnet E, Guipponi M, Vannier A, Radovanovic I, Bena F, Forestier F et al (2014) Extrachromosomal driver mutations in glioblastoma and low-grade glioma. Nat Commun 5:5690. https://doi.org/10.1038/ncomms6690 PubMedPubMedCentralCrossRefGoogle Scholar
- 159.Noushmehr H, Weisenberger DJ, Diefes K, Phillips HS, Pujara K, Berman BP, Pan F, Pelloski CE, Sulman EP, Bhat KP et al (2010) Identification of a CpG island methylator phenotype that defines a distinct subgroup of glioma. Cancer Cell 17:510–522. https://doi.org/10.1016/j.ccr.2010.03.017 PubMedPubMedCentralCrossRefGoogle Scholar
- 166.Ostrom QT, Gittleman H, Fulop J, Liu M, Blanda R, Kromer C, Wolinsky Y, Kruchko C, Barnholtz-Sloan JS (2015) CBTRUS statistical report: primary brain and central nervous system tumors diagnosed in the United States in 2008–2012. Neurooncology 17(Suppl 4):iv1–iv62 https://doi.org/10.1093/neuonc/nov189
- 168.Ozawa T, Riester M, Cheng YK, Huse JT, Squatrito M, Helmy K, Charles N, Michor F, Holland EC (2014) Most human non-GCIMP glioblastoma subtypes evolve from a common proneural-like precursor glioma. Cancer Cell 26:288–300. https://doi.org/10.1016/j.ccr.2014.06.005 PubMedPubMedCentralCrossRefGoogle Scholar
- 171.Patel AP, Tirosh I, Trombetta JJ, Shalek AK, Gillespie SM, Wakimoto H, Cahill DP, Nahed BV, Curry WT, Martuza RL et al (2014) Single-cell RNA-seq highlights intratumoral heterogeneity in primary glioblastoma. Science 344:1396–1401. https://doi.org/10.1126/science.1254257 PubMedPubMedCentralCrossRefGoogle Scholar
- 172.Pekmezci M, Rice T, Molinaro AM, Walsh KM, Decker PA, Hansen H, Sicotte H, Kollmeyer TM, McCoy LS, Sarkar G et al (2017) Adult infiltrating gliomas with WHO 2016 integrated diagnosis: additional prognostic roles of ATRX and TERT. Acta Neuropathol 133:1001–1016. https://doi.org/10.1007/s00401-017-1690-1 PubMedCrossRefPubMedCentralGoogle Scholar
- 175.Perry A, Wesseling P (2016) Histologic classification of gliomas. Handb Clin Neurol 134:71–95. https://doi.org/10.1016/B978-0-12-802997-8.00005-0 PubMedCrossRefGoogle Scholar
- 177.Raabe EH, Lim KS, Kim JM, Meeker A, Mao XG, Nikkhah G, Maciaczyk J, Kahlert U, Jain D, Bar E et al (2011) BRAF activation induces transformation and then senescence in human neural stem cells: a pilocytic astrocytoma model. Clin Cancer Res 17:3590–3599. https://doi.org/10.1158/1078-0432.CCR-10-3349 PubMedPubMedCentralCrossRefGoogle Scholar
- 181.Reuss DE, Sahm F, Schrimpf D, Wiestler B, Capper D, Koelsche C, Schweizer L, Korshunov A, Jones DT, Hovestadt V et al (2015) ATRX and IDH1-R132H immunohistochemistry with subsequent copy number analysis and IDH sequencing as a basis for an “integrated” diagnostic approach for adult astrocytoma, oligodendroglioma and glioblastoma. Acta Neuropathol 129:133–146. https://doi.org/10.1007/s00401-014-1370-3 PubMedCrossRefGoogle Scholar
- 184.Rouleau GA, Merel P, Lutchman M, Sanson M, Zucman J, Marineau C, Hoang-Xuan K, Demczuk S, Desmaze C, Plougastel B et al (1993) Alteration in a new gene encoding a putative membrane-organizing protein causes neuro-fibromatosis type 2. Nature 363:515–521. https://doi.org/10.1038/363515a0 PubMedCrossRefGoogle Scholar
- 186.Sanborn JZ, Salama SR, Grifford M, Brennan CW, Mikkelsen T, Jhanwar S, Katzman S, Chin L, Haussler D (2013) Double minute chromosomes in glioblastoma multiforme are revealed by precise reconstruction of oncogenic amplicons. Can Res 73:6036–6045. https://doi.org/10.1158/0008-5472.CAN-13-0186 CrossRefGoogle Scholar
- 188.Sasaki M, Knobbe CB, Itsumi M, Elia AJ, Harris IS, Chio II, Cairns RA, McCracken S, Wakeham A, Haight J et al (2012) D-2-hydroxyglutarate produced by mutant IDH1 perturbs collagen maturation and basement membrane function. Gene Dev 26:2038–2049. https://doi.org/10.1101/gad.198200.112 PubMedPubMedCentralCrossRefGoogle Scholar
- 189.Schiffman JD, Hodgson JG, VandenBerg SR, Flaherty P, Polley MY, Yu M, Fisher PG, Rowitch DH, Ford JM, Berger MS et al (2010) Oncogenic BRAF mutation with CDKN2A inactivation is characteristic of a subset of pediatric malignant astrocytomas. Can Res 70:512–519. https://doi.org/10.1158/0008-5472.CAN-09-1851 CrossRefGoogle Scholar
- 190.Schindler G, Capper D, Meyer J, Janzarik W, Omran H, Herold-Mende C, Schmieder K, Wesseling P, Mawrin C, Hasselblatt M et al (2011) Analysis of BRAF V600E mutation in 1320 nervous system tumors reveals high mutation frequencies in pleomorphic xanthoastrocytoma, ganglioglioma and extra-cerebellar pilocytic astrocytoma. Acta Neuropathol 121:397–405. https://doi.org/10.1007/s00401-011-0802-6 PubMedCrossRefGoogle Scholar
- 191.Schwartzentruber J, Korshunov A, Liu XY, Jones DT, Pfaff E, Jacob K, Sturm D, Fontebasso AM, Quang DA, Tonjes M et al (2012) Driver mutations in histone H3.3 and chromatin remodelling genes in paediatric glioblastoma. Nature 482:226–231. https://doi.org/10.1038/nature10833 PubMedCrossRefGoogle Scholar
- 197.Solimini NL, Xu Q, Mermel CH, Liang AC, Schlabach MR, Luo J, Burrows AE, Anselmo AN, Bredemeyer AL, Li MZ et al (2012) Recurrent hemizygous deletions in cancers may optimize proliferative potential. Science 337:104–109. https://doi.org/10.1126/science.1219580 PubMedPubMedCentralCrossRefGoogle Scholar
- 198.Sottoriva A, Spiteri I, Piccirillo SG, Touloumis A, Collins VP, Marioni JC, Curtis C, Watts C, Tavare S (2013) Intratumor heterogeneity in human glioblastoma reflects cancer evolutionary dynamics. Proc Natl Acad Sci USA 110:4009–4014. https://doi.org/10.1073/pnas.1219747110 PubMedPubMedCentralCrossRefGoogle Scholar
- 199.Steck PA, Pershouse MA, Jasser SA, Yung WK, Lin H, Ligon AH, Langford LA, Baumgard ML, Hattier T, Davis T et al (1997) Identification of a candidate tumour suppressor gene, MMAC1, at chromosome 10q23.3 that is mutated in multiple advanced cancers. Nat Genet 15:356–362. https://doi.org/10.1038/ng0497-356 PubMedCrossRefGoogle Scholar
- 202.Sturm D, Witt H, Hovestadt V, Khuong-Quang DA, Jones DT, Konermann C, Pfaff E, Tonjes M, Sill M, Bender S et al (2012) Hotspot mutations in H3F3A and IDH1 define distinct epigenetic and biological subgroups of glioblastoma. Cancer Cell 22:425–437. https://doi.org/10.1016/j.ccr.2012.08.024 PubMedCrossRefGoogle Scholar
- 203.Suva ML, Rheinbay E, Gillespie SM, Patel AP, Wakimoto H, Rabkin SD, Riggi N, Chi AS, Cahill DP, Nahed BV et al (2014) Reconstructing and reprogramming the tumor-propagating potential of glioblastoma stem-like cells. Cell 157:580–594. https://doi.org/10.1016/j.cell.2014.02.030 PubMedPubMedCentralCrossRefGoogle Scholar
- 207.Tarabichi M, Martincorena I, Gerstung M, Markowetz F, Spellman PT, Morris QD, Lingjaerde OC, Wedge DC, Van Loo P (2017) Neutral tumor evolution? bioRxiv: https://doi.org/10.1101/158006
- 209.Telomeres Mendelian Randomization C, Haycock PC, Burgess S, Nounu A, Zheng J, Okoli GN, Bowden J, Wade KH, Timpson NJ, Evans DM et al (2017) Association between telomere length and risk of cancer and non-neoplastic diseases: a mendelian randomization study. JAMA Oncol 3:636–651. https://doi.org/10.1001/jamaoncol.2016.5945 CrossRefGoogle Scholar
- 210.Tirosh I, Venteicher AS, Hebert C, Escalante LE, Patel AP, Yizhak K, Fisher JM, Rodman C, Mount C, Filbin MG et al (2016) Single-cell RNA-seq supports a developmental hierarchy in human oligodendroglioma. Nature 539:309–313. https://doi.org/10.1038/nature20123 PubMedPubMedCentralCrossRefGoogle Scholar
- 215.Uhrbom L, Dai C, Celestino JC, Rosenblum MK, Fuller GN, Holland EC (2002) Ink4a-Arf loss cooperates with KRas activation in astrocytes and neural progenitors to generate glioblastomas of various morphologies depending on activated Akt. Can Res 62:5551–5558Google Scholar
- 218.van Thuijl HF, Scheinin I, Sie D, Alentorn A, van Essen HF, Cordes M, Fleischeuer R, Gijtenbeek AM, Beute G, van den Brink WA et al (2014) Spatial and temporal evolution of distal 10q deletion, a prognostically unfavorable event in diffuse low-grade gliomas. Genome Biol 15:471. https://doi.org/10.1186/s13059-014-0471-6 PubMedPubMedCentralCrossRefGoogle Scholar
- 219.Venteicher AS, Tirosh I, Hebert C, Yizhak K, Neftel C, Filbin MG, Hovestadt V, Escalante LE, Shaw ML, Rodman C et al (2017) Decoupling genetics, lineages, and microenvironment in IDH-mutant gliomas by single-cell RNA-seq. Science. https://doi.org/10.1126/science.aai8478 PubMedPubMedCentralCrossRefGoogle Scholar
- 220.Verhaak RG, Hoadley KA, Purdom E, Wang V, Qi Y, Wilkerson MD, Miller CR, Ding L, Golub T, Mesirov JP et al (2010) Integrated genomic analysis identifies clinically relevant subtypes of glioblastoma characterized by abnormalities in PDGFRA, IDH1, EGFR, and NF1. Cancer Cell 17:98–110. https://doi.org/10.1016/j.ccr.2009.12.020 PubMedPubMedCentralCrossRefGoogle Scholar
- 226.Walsh KM, Codd V, Rice T, Nelson CP, Smirnov IV, McCoy LS, Hansen HM, Elhauge E, Ojha J, Francis SS et al (2015) Longer genotypically-estimated leukocyte telomere length is associated with increased adult glioma risk. Oncotarget 6:42468–42477. https://doi.org/10.18632/oncotarget.6468 PubMedPubMedCentralCrossRefGoogle Scholar
- 227.Walsh KM, Codd V, Smirnov IV, Rice T, Decker PA, Hansen HM, Kollmeyer T, Kosel ML, Molinaro AM, McCoy LS et al (2014) Variants near TERT and TERC influencing telomere length are associated with high-grade glioma risk. Nat Genet 46:731–735. https://doi.org/10.1038/ng.3004 PubMedPubMedCentralCrossRefGoogle Scholar
- 230.Wang Q, Hu B, Hu X, Kim H, Squatrito M, Scarpace L, deCarvalho AC, Lyu S, Li P, Li Y et al (2017) Tumor evolution of Glioma-Intrinsic gene expression subtypes associates with immunological changes in the microenvironment. Cancer Cell 32(42–56):e46. https://doi.org/10.1016/j.ccell.2017.06.003 CrossRefGoogle Scholar
- 231.Ward PS, Patel J, Wise DR, Abdel-Wahab O, Bennett BD, Coller HA, Cross JR, Fantin VR, Hedvat CV, Perl AE et al (2010) The common feature of leukemia-associated IDH1 and IDH2 mutations is a neomorphic enzyme activity converting alpha-ketoglutarate to 2-hydroxyglutarate. Cancer Cell 17:225–234. https://doi.org/10.1016/j.ccr.2010.01.020 PubMedPubMedCentralCrossRefGoogle Scholar
- 234.Wiestler B, Capper D, Hovestadt V, Sill M, Jones DT, Hartmann C, Felsberg J, Platten M, Feiden W, Keyvani K et al (2014) Assessing CpG island methylator phenotype, 1p/19q codeletion, and MGMT promoter methylation from epigenome-wide data in the biomarker cohort of the NOA-04 trial. Neuro-oncology 16:1630–1638. https://doi.org/10.1093/neuonc/nou138 PubMedPubMedCentralCrossRefGoogle Scholar
- 238.Wrensch M, Jenkins RB, Chang JS, Yeh RF, Xiao Y, Decker PA, Ballman KV, Berger M, Buckner JC, Chang S et al (2009) Variants in the CDKN2B and RTEL1 regions are associated with high-grade glioma susceptibility. Nat Genet 41:905–908. https://doi.org/10.1038/ng.408 PubMedPubMedCentralCrossRefGoogle Scholar
- 241.Wu G, Broniscer A, McEachron TA, Lu C, Paugh BS, Becksfort J, Qu C, Ding L, Huether R, Parker M et al (2012) Somatic histone H3 alterations in pediatric diffuse intrinsic pontine gliomas and non-brainstem glioblastomas. Nat Genet 44:251–253. https://doi.org/10.1038/ng.1102 PubMedPubMedCentralCrossRefGoogle Scholar
- 243.Xu W, Yang H, Liu Y, Yang Y, Wang P, Kim SH, Ito S, Yang C, Wang P, Xiao MT et al (2011) Oncometabolite 2-hydroxyglutarate is a competitive inhibitor of alpha-ketoglutarate-dependent dioxygenases. Cancer Cell 19:17–30. https://doi.org/10.1016/j.ccr.2010.12.014 PubMedPubMedCentralCrossRefGoogle Scholar
- 246.Zhang J, Wu G, Miller CP, Tatevossian RG, Dalton JD, Tang B, Orisme W, Punchihewa C, Parker M, Qaddoumi I et al (2013) Whole-genome sequencing identifies genetic alterations in pediatric low-grade gliomas. Nat Genet 45:602–612. https://doi.org/10.1038/ng.2611 PubMedPubMedCentralCrossRefGoogle Scholar
- 247.Zheng S, Fu J, Vegesna R, Mao Y, Heathcock LE, Torres-Garcia W, Ezhilarasan R, Wang S, McKenna A, Chin L et al (2013) A survey of intragenic breakpoints in glioblastoma identifies a distinct subset associated with poor survival. Genes Dev 27:1462–1472. https://doi.org/10.1101/gad.213686.113 PubMedPubMedCentralCrossRefGoogle Scholar
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.